- Email: [email protected]
Association between MDR1 C3435T polymorphism and refractory epilepsy in the Chinese population: A systematic review and meta-analysis
Epilepsy & Behavior 36 (2014) 173–179
Contents lists available at ScienceDirect
Epilepsy & Behavior journal homepage: www.elsevier.com/locate/yebeh
Association between MDR1 C3435T polymorphism and refractory epilepsy in the Chinese population: A systematic review and meta-analysis Ji-Wei Cheng a, Li-Jun Zhang a, Yu-Qing Hou a, Qing Zhao a, Xiao-Jing Zhang a, Xue-Fen Chen b, Yu Bai a,⁎ a b
Department of Neurology, Putuo Hospital, Shanghai University of Traditional Chinese Medicine, Shanghai 200062, China Department of Preventive Medicine and Health Statistics of Shanghai University of Traditional Chinese Medicine, Shanghai 201203, China
a r t i c l e
i n f o
Article history: Received 29 January 2014 Revised 8 April 2014 Accepted 6 May 2014 Available online xxxx Keywords: MDR1 C3435T Polymorphism Epilepsy Chinese Meta-analysis
a b s t r a c t The association between the C3435T polymorphism in the MDR1 gene and refractory epilepsy remains controversial. The association appears to be influenced by ethnicity and region. We have performed a systematic review and meta-analysis to assess the link between the MDR1 C3435T polymorphism and refractory epilepsy in the Chinese population. We searched the Cochrane Library, MIDLINE, EMBASE, CBM disc, CNKI, VIP, and WANFANG databases for literature published through August 2013 for case–control studies that evaluated the association between the MDR1 C3435T polymorphism and refractory epilepsy. Twenty-one case–control studies involving 4269 patients (1863 cases in the group with drug-resistant epilepsy and 2406 in the group with drug-responsive epilepsy) were included in the systematic review and metaanalysis. The analysis showed that there were significantly more cases with the MDR1 3435 CC genotype in the group with drug-resistant epilepsy than in the group with drug-responsive epilepsy [odds ratio (OR) = 1.50, 95% confidence interval (CI) = 1.09–2.06, P = 0.01]. In a subanalysis of patients from the southern regions of China, the correlation was not significant [odds ratio (OR) = 1.2, 95% confidence interval (CI) = 0.89–1.64, P = 0.24]. The relationship established in a subset of the Chinese population between the MDR1 C3435T polymorphism and refractory epilepsy will guide epilepsy treatment and development of new AEDs. © 2014 Elsevier Inc. All rights reserved.
1. Introduction Approximately 50 million people worldwide suffer from epilepsy, and roughly one-third cannot control the disease despite treatment with appropriate antiepileptic drugs (AEDs) [1,2]. China has about 9 million people with epilepsy [3]. A number of studies have demonstrated a relationship between refractory epilepsy and the multidrug resistance gene 1 (MDR1/ABCB1) [4–7]. MDR1 is highly polymorphic, and the C3435T polymorphism is one of the most common [8]. Although the C3435T polymorphism is associated with refractory epilepsy [4, 9–11], there are indications that the link is specific to only certain populations. The studies of Siddiqui et al. [4] and Sayyah et al. [11] found that patients with drug-resistant epilepsy were more likely to have a CC genotype at MDR1 3435 than a TT genotype. However, Seo et al. [9] and Kwan et al. [12] found that patients with drug-resistant epilepsy were more likely to have TT genotypes. Other studies [2,13–15] have found no relation between the MDR1 sequence and drug resistance. It has been suggested that these differences may be related to ethnicity
⁎ Corresponding author. Tel./fax: +86 13818606517. E-mail address: [email protected] (Y. Bai).
http://dx.doi.org/10.1016/j.yebeh.2014.05.007 1525-5050/© 2014 Elsevier Inc. All rights reserved.
and region [16,17]. We have retrieved relevant studies published in China and abroad and have carried out a systematic review and metaanalysis on the link between the C3435T polymorphism and refractory epilepsy in the Chinese population. 2. Methods 2.1. Search strategy and selection We searched the Cochrane Library, MIDLINE, EMBASE, CBM disc, CNKI, VIP, and WANFANG databases for case–control or cohort studies published through August of 2013 that evaluated the association between polymorphisms of MDR1 C3435T and refractory epilepsy in Chinese populations, region and language unlimited. The case–control studies included in this meta-analysis were required to meet the following criteria: (a) patients were treated with AEDs, and compliance was demonstrated; (b) data on genotype distributions were available for both case and control groups; (c) genotype distribution of the control subjects conformed to the Hardy–Weinberg equilibrium; and (d) there was a clear treatment outcome of either resistance or responsiveness to AEDs. The Boolean search strategy “C3435T” AND “[MDR1 OR ABCB1 OR P-glycoprotein (P-gp)]” AND “epilepsy” AND “Chinese” was
Various Various Various CBZ Various Various VPA VPA/CBZ Various Various Various
Various Various Various Various Various Various Various Various Various Various
76 178 61 93 12 152 20 33 58 35 T 59 27 45 33 101 84 86 115 111 57 22
Abbreviations: R, drug-responsive; NR, drug-resistant; MDR1, multidrug resistance 1. VPA, valproate; CBZ, carbamazepine; “various” indicates that the patient was treated with two or more drugs.
Sex (m) / / 32 / / / 33 43 54 32 Sex (m) / 25 / / 67 / 56 / 108 47 19 Age 38.5 ± 13.4 / 29.39 ± 14.71 / / / 27.9 ± 12.1 27 ± 13 37.33 ± 15.56 36.51 ± 13.50 Age / 5.3 ± 1.6 1.5–15 16–73 5–59 / 30 ± 19 / 6.6 ± 3.9 47.33 ± 20.39 21.2 ± 17.5 T 266 205 69 241 22 190 42 68 102 64 T 45 36 123 40 123 66 89 125 139 64 25 C 180 389 55 185 46 350 24 82 78 56 C 79 44 205 58 121 76 105 167 247 94 27 TT 74 22 16 67 1 21 15 12 29 19 TT 8 8 22 11 19 14 22 28 28 17 5 CT 118 161 37 107 20 148 12 44 44 26 CT 29 20 79 18 85 38 45 69 83 30 15
R group (genotype/allele/age/sex)
CC 31 114 9 39 13 101 6 19 17 15 CC 25 12 63 20 18 19 30 49 82 32 6 Sex (m) / / 51 / / / 19 31 41 30 Sex (m) / 23 / / 62 / 56 / 95 54 18 Age 39.1 ± 11 / 32.58 ± 15.41 / / / 33.8 ± 13.8 29 ± 12 36.74 ± 11.63 35.28 ± 13.20 Age / 5.7 ± 1.3 1.5–15 16–73 5–55 / 29 ± 18 / 6.2 ± 3.6 30.96 ± 15.18 31.3 ± 16.5 T
C 140 264 83 135 10 236 42 103 132 91 C 81 53 55 29 115 86 114 147 203 125 24 TT 15 37 13 19 2 29 9 10 5 6 TT 13 4 10 9 18 20 16 23 18 10 4 China (Shanghai) China (Zhejiang) China (Shenzhen/Hong Kong) China (Jilin) China (Shandong) China (Fujian) Malaya (Kuala Lumpur) Malaya (Kuala Lumpur) China (Sichuan) China (Nanjing) China (Shenzhen)
In seven of the 21 published articles, the authors concluded that patients with epilepsy with the MDR1 3435 CC genotype were more likely to have drug-resistant epilepsy compared with those with a different genotype. Three studies identified an association between the 3435 TT genotype and drug resistance; however, 11 studies found no such link. Our meta-analysis showed that there were more MDR1 3435 CC genotype patients in the group with drug-resistant epilepsy than in the group with drug-responsive epilepsy (Fig. 1 and Table 2). Our analysis indicated no significant association between the CT or TT genotypes and resistance to AEDs (Figs. 2 and 3 and Table 2). Studies included in
China (Taiwan) China (Hong Kong) China (Shanghai) China (Taiwan) China (Hong Kong) China (Hong Kong) China (Tianjin) China (Beijing) China (Ningxia) China (Ningxia)
3.2. Meta-analysis
CT 46 104 35 55 8 94 2 13 48 23 CT 33 19 25 15 65 44 54 69 75 32 14
Twenty-two published articles were retrieved that evaluated the relationship between the MDR1 polymorphisms and refractory epilepsy in Chinese patients. Of these articles, eight [2,10,12,18–22] were published in English, and 14 [23–35] were published in Chinese. One of these was a published dissertation (written in Chinese). All but the dissertation were included in the systematic review and meta-analysis. The results of each included study are summarized in Table 1. Six studies were conducted in northern China, 13 studies in southern China, and two in Malaysia (included in the South region for subanalysis). A total of 4269 patients were included in the meta-analysis; 1863 cases had drug-resistant epilepsy, and 2406 cases had drug-responsive epilepsy. There were no significant difference in ages (WMD = − 0.18, 95% CI = − 1.83 to 1.46, P = 0.83) or genders (OR = 0.98, 95% CI = 0.81 to 1.19, P = 0.86) between the two groups.
CC 47 80 24 40 1 71 20 45 42 34 CC 24 17 15 7 25 21 30 39 64 44 5
3.1. Study characteristics
NR group (genotype/allele/age/sex)
3. Results
Countries (districts)
The RevMan 4.2 software was used for the meta-analysis. The odds ratios (ORs) and 95% confidence intervals (CIs) of the binary classification variable data and the weighted mean differences (WMDs) and 95% CIs of the continuous variable data were calculated using a fixed effects model if there was no significant heterogeneity between each test (P N 0.05, I2 b 50%). If heterogeneity was present, a random effects model was employed. Twenty-one studies were analyzed. The study populations were divided for subanalyses into patients from two regions to explore the influence of region: the North (northern China) and the South (including southern China and Malaysia) as separated by the Yangtze River. We also performed a sensitivity analysis by eliminating each study in turn and a cumulative analysis by literature publication dates to test the stability of our conclusions and identify the sources of heterogeneity. Publication bias was evaluated by visual funnel plot inspection and Egger's test.
Positive studies (n = 10) Hung et al. [18] Kwan et al. [12] Lu et al. [23] Hung et al. [19] Szoeke et al. [2] Kwan et al. [10] Zheng et al. [24] Ding et al. [25] Dong et al. [22] Dong et al. [26] Negative studies (n = 11) Gao et al. [27] Wang et al. [28] Chen et al. [29] Guo et al. [30] Jin et al. [31] Wang et al. [32] Haerian et al. [20] Haerian et al. [21] Dong et al. [35] Di et al. [33] Yang et al. [34]
2.3. Statistical analysis
Table 1 Characteristics of the studies of the MDR1 C3435T polymorphism and the distribution of the MDR1 C3435T genotypes and allele frequencies among patients whose epilepsy is sensitive (NR) or resistant (R) to AEDs.
We downloaded and printed the full text of all relevant publications. Two trained valuators read the literature and extracted the relevant information into registration forms according to a standardized method. Each publication was read and data were extracted by both valuators. The information included the name, gender, and age of first author of the study, year of publication, sample size, the country (district) of the study, and the research conclusions. Disagreements were documented and resolved by discussion by a third author.
Drugs
2.2. Data extraction
CC TT CC CC TT TT CC CC CC CC
Frequent genotype in NR
used. We also retrieved the related references listed in the studies included in our meta-analysis. All studies included were published literature; dissertations and conference papers were not included.
– – – – – – – – – – –
J.-W. Cheng et al. / Epilepsy & Behavior 36 (2014) 173–179
Studies
174
J.-W. Cheng et al. / Epilepsy & Behavior 36 (2014) 173–179
175
Fig. 1. Forest plot of MDR1 CC genotype frequencies in meta-analysis of patients with drug-resistant and drug-responsive epilepsies.
the meta-analysis were heterogeneous, so we adopted the random effects model. Visual funnel plot inspection and Egger's test showed no obvious publication biases (CC, P = 0.49; CT, P = 0.64; TT, P = 0.80). We conducted a region-based subanalysis, sensitivity analysis, and cumulative meta-analysis by literature publication date to test the stability of our conclusions and to identify the sources of heterogeneity. The subanalysis showed that in patients from the southern region of China, which included Malaysia, there was no significant association between the MDR1 3435 CC genotype and drug resistance (Fig. 1 and Table 2). The subanalysis of the patients from the North was consistent with the overall results that the 3435 CC genotype was correlated with drug resistance (Table 2).
We performed a sensitivity analysis by eliminating each study in turn. The heterogeneity of the North subanalysis on 3435 CC disappeared in the absence of the study of Guo et al. [30]. The heterogeneity of the North subanalysis and of the total analysis on 3435 CT decreased when data from the study of Ding et al. [25] were removed. After eliminating the study of Dong et al. [26], the heterogeneity of the North subanalysis on 3435 TT disappeared. After eliminating the studies of Hung et al. [18,19], the heterogeneity of the South subanalysis on 3435 CC disappeared. We also conducted a cumulative meta-analysis on the association between the MDR1 3435 CC genotype and drug resistance by literature publication date; the results were in agreement with those obtained when all data were analyzed (Fig. 4).
Table 2 Distribution of MDR1 C3435T genotype frequencies among patients with drug-resistant and drug-responsive epilepsies based on district within China. Genotypes
Studies
CC
T N S T N S T N S
TT
CT
n/N
Fixed-effects model
NR
R
695/1863 173/396 522/1467 290/1863 57/396 233/1467 873/1863 166/396 707/1467
720/2406 95/429 625/1977 458/2406 105/429 353/1977 1228/2406 229/429 999/1977
P
0.22
OR (95% CI)
0.92 (0.80–1.05)
Random-effects model P
OR (95% CI)
0.01 0.01 0.24 0.16 0.10 0.61 0.10 0.10
1.50 (1.09–2.06) 2.75 (1.36–5.55) 1.20 (0.89–1.64) 0.78 (0.54–1.13) 0.52 (0.24–1.13) 0.91 (0.62–1.32) 0.85 (0.71–1.03) 0.57 (0.29–1.11)
I2 (%)
P
79.90 78.30 73.40 73.00 75.40 71.70 49.50 78.70 0.00
0.00 0.00 0.00 0.00 0.00 0.00 0.01 0.00 0.64
Abbreviations: R, drug-responsive; NR, drug-resistant; T, includes all studies; N, includes only studies on northern China; S, includes studies on southern China and Malaysia.
176
J.-W. Cheng et al. / Epilepsy & Behavior 36 (2014) 173–179
Fig. 2. Forest plot of MDR1 CT genotype frequencies in meta-analysis of patients with drug-resistant and drug-responsive epilepsies.
4. Discussion Drug resistance is mainly mediated by the protein products of the MDR1 and MDR2 genes. P-glycoprotein, which is the expression product of MDR1, is expressed in the intestine, placenta, blood–brain barrier, kidney, and liver. The substrates of P-gp include liposoluble drugs. Most AEDs, including phenytoin, phenobarbital, and lamotrigine, are liposoluble. In patients whose epilepsy is resistant to AEDs, it is assumed that P-gp actively transports these AEDs from cells [36,37]. In humans, MDR1 is located on chromosome 7 q21.1, and more than 50 single nucleotide polymorphisms (SNPs) have been identified. There is an association between C3435T of exon 26 of the MDR1 gene and the expression and function of P-gp [38]. Siddiqui et al. [4] were first to report that patients with drugresistant epilepsy were more likely to have the CC genotype at MDR1 3435 than the TT genotype. Results of subsequent studies have been inconsistent. Data from some studies [11,39] have supported this association, whereas others [18,19] found that patients with the TT genotype were more likely to have drug-resistant epilepsy, and, still, other studies [2,13–15,21,22,40] failed to find any association between this position of MDR1 and drug resistance in patients with epilepsy. There are a number of potential causes for these contradictory results. First, a uniform definition of refractory epilepsy was not used in previous analyses, making division of patients into those with drug-resistant and those with drug-responsive epilepsy difficult [1,4]. As listed in Table 3, different definitions of drug resistance were used in the studies analyzed in our meta-analysis. In the studies of Hung et al. [18,19], drug
resistance was defined as 10 seizures in a year with two or more AEDs at maximally tolerated doses. Kwan et al. [12] defined it as one seizure in a month with two or more AEDs at therapeutic dosages, and Haerian et al. [21] defined it as one seizure in a year with valproate or carbamazepine monotherapy at maximally tolerated dosages. In contrast, most of the studies of Chinese mainland subjects [23–26,28,30–32,34,35] defined drug resistance as less than 50% seizure reduction with AEDs at maximally tolerated doses or more than four seizures in a month with appropriate first-line AEDs at therapeutic dosages for 2 years. These differences made it difficult to group patients according to the same criterion. Second, the types and doses of AEDs vary from study to study. Certain antiepileptic drugs, such as phenobarbital, are P-gp substrates, whereas others, such as sodium valproate, are not. Other drugs that might be taken by patients with epilepsy, including calcium channel blockers and antibiotics, are also P-gp substrates and might have confounded results. Third, the frequency of the C3435T polymorphism varies with ethnicity in healthy subjects. Those of African descent are most likely to have the 3435 CC genotype with Europeans least likely [16,17,41]. Fourth, age and gender may impact results. The study of Sanchez et al. on adults with epilepsy found that 3435 CC genotype patients were more likely to have drug-resistant epilepsy compared with those with a different genotype; however, this was not the case for children. Sayyah et al. [11] found that female, but not male, patients in Iran with the 3435 CC genotype were more likely to have drug-resistant epilepsy compared with those with other genotypes. Fifth, differences in genotype analysis methods may also affect the results: Lakhan et al. [40] detected a 231-bp PCR product using DpnII restriction endonuclease,
J.-W. Cheng et al. / Epilepsy & Behavior 36 (2014) 173–179
177
Fig. 3. Forest plot of MDR1 TT genotype frequencies in meta-analysis of patients with drug-resistant and drug-responsive epilepsies.
Fig. 4. Cumulative meta-analysis of the relationship between MDR1 3435 CC and drug resistance by study publication date.
whereas Lu et al. [23] detected a 248-bp PCR product using DpnI. Finally, the etiopathogenesis of refractory epilepsy is complex. It appears to be associated with excessive expression of P-gp in the brain and may be affected by CY450, interleukin, and other enzymes that alter the metabolism of AEDs. Meta-analysis can increase the reliability of conclusions by increasing sample size. Data from 21 studies in which drug resistance or sensitivity and genotypes of patients with epilepsy from China were determined were included in our meta-analysis. These case–control studies involved 4269 patients. The meta-analysis showed that a significantly higher fraction of patients in the group with drug-resistant epilepsy than in the group with drug-sensitive epilepsy had the MDR1 3435 CC genotype. However, two previous meta-analyses did not support the association [41,42]. In the study of Bournissen et al. [41], 11 studies from six countries around the world involving 3371 patients were included in the analysis. We found four positive and seven negative studies in that analysis. The most frequent genotype in the group with drug-resistant epilepsy from only one of these studies was 3435 CC. The meta-analysis found no association between the ABC1 genotype and response to AEDs. In the meta-analysis of Haerian et al. [42], 22 studies from 11 countries around the world involving 6755 patients were included in the analysis. There were nine positive and 13 negative studies in the analysis. The most frequent genotype in the group with drug-resistant epilepsy from five of the studies was 3435 CC and
178
J.-W. Cheng et al. / Epilepsy & Behavior 36 (2014) 173–179
Table 3 Definitions of drug responsiveness and drug resistance used in analyzed studies. Studies
Countries (districts)
Definition of drug responsiveness and drug resistance Drug responsiveness
Drug resistance N10 seizures in a year with 2 or more AEDs at maximally tolerated doses N1 seizure in a month with 2 or more AEDs at therapeutic dosages
Positive studies Hung et al. [18] Kwan et al. [12]
China (Taiwan) China (Hong Kong)
Lu et al. [23] Hung et al. [19] Szoeke et al. [2] Kwan et al. [10] Zheng et al. [24]
China (Shanghai) China (Taiwan) China (Hong Kong) China (Hong Kong) China (Tianjin)
N2 years seizure freedom N1 year seizure freedom with a stable dose of an AED ≥50% seizure reduction As Hung et al. [18] N1 year seizure freedom As Kwan et al. [12] N1 year seizure freedom
Ding et al. [25] Dong et al. [22] Dong et al. [26]
China (Beijing) China (Ningxia) China (Ningxia)
≥50% seizure reduction ≥50% seizure reduction ≥50% seizure reduction
Negative studies Gao et al. [27]
China (Shanghai)
≥1 year seizure freedom
Wang et al. [28]
China (Zhejiang)
N1 year seizure freedom
Chen et al. [29] Guo et al. [30]
China (Shenzhen/Hong Kong) China (Jilin)
Jin et al. [31]
China (Shandong)
Wang et al. [32]
China (Fujian)
N1 year seizure freedom ≥50% seizure reduction with CBZ monotherapy N1 year seizure freedom or ≥50% seizure reduction N1 year seizure freedom
Haerian et al. [20]
Malaya (Kuala Lumpur)
Haerian et al. [21]
Malaya (Kuala Lumpur)
Dong et al. [35] Di et al. [33] Yang et al. [34]
China (Sichuan) China (Nanjing) China (Shenzhen)
N1 year seizure freedom with VPA treatment N1 year seizure freedom with CBZ or VPA monotherapy N1 year seizure freedom N1 year seizure freedom N1 year seizure freedom
3435TT from four of the studies. The meta-analysis did not find an association between the ABC1 C3435T polymorphism and the risk of resistance to AEDs. The main difference between the two previous meta-analyses and the present study is the case ethnicities. Our study concentrated on a Chinese population, whereas cases analyzed in the previous studies were from around the world. There is precedence for our finding that ethnicity is a factor in genetic linkage of diseases. Wang et al. [43] performed a meta-analysis with ethnically heterogeneous groups from all over the world including thirty-nine studies that included 6863 cases with gastric cancer and 8434 controls. The analysis suggested that the IL-1RN genetic polymorphisms were associated with an increased risk of developing gastric cancer. A subanalysis of the Chinese population did not find such an association. The meta-analysis performed by Li et al. [44] of nineteen case–control studies involving 2223 patients and 2936 controls suggested that the MTHFR C667T genetic polymorphism was significantly associated with increased risk of ischemic stroke. In contrast, the study of Marie et al. [45] found that the MTHFR gene polymorphism was not related to the risk of ischemic cerebrovascular disease in the Brazilian population. In order to test the stability of our conclusion, we divided our studies into subgroups of patients from the north of China to the south of China (including Malaysia) and conducted a cumulative meta-analysis by literature publication date. The results, with the exception of subanalysis of the data from the South, were much the same as those obtained when all data were analyzed simultaneously. Heterogeneity inspection must be conducted in meta-analysis as only the statistics of homogeneous data are capable of a merger. The heterogeneity of data used here was examined by χ2 and I2 tests. These tests suggested
b50% of seizure reduction with AEDs at maximally tolerated doses As Hung et al. [18] Seizure in a year As Kwan et al. [12] N4 seizures in a month with appropriate first-line AEDs at therapeutic dosages for 2 years b50% seizure reduction with AEDs at maximally tolerated doses b50% seizure reduction with AEDs at maximally tolerated doses b50% seizure reduction with AEDs at maximally tolerated doses N1 seizure in a month with 2 or more first-line AEDs at therapeutic dosages for 6 months N4 seizures in a month with appropriate first-line AEDs at therapeutic dosages for 2 years N4 seizures in a year b50% seizure reduction with CBZ monotherapy b50% seizure reduction with 2 or more AEDs at therapeutic dosages N4 seizures in a month with appropriate first-line AEDs at therapeutic dosages for 2 years N1 seizure in a year with VPA at maximally tolerated doses N1 seizure in a year with CBZ or VPA monotherapy at maximally tolerated dosages N4 seizures in a year ≥1 seizure in a month with 2 or more appropriate AEDs for 2 years N4 seizures in a month with appropriate first-line AEDs at therapeutic dosages for 2 years
heterogeneity, and, therefore, we adopted the random effects model. We also eliminated data from each study in turn to find the source of the heterogeneity. The heterogeneity decreased upon elimination of certain studies, but the removal of data from individual studies had no impact on the overall conclusion. Heterogeneity may be associated with differences in sample size. Visual funnel plot inspection and Egger's test did not show obvious publication bias. It is commonly believed that valproic acid is not a substrate of p-glycoprotein but a substrate of multidrug resistance-associated protein [46]. However, VPA was the most commonly used firstline AED in reports that demonstrated both negative [20,21,47,48] and positive [49] associations between polymorphisms and drug efficacy. Furthermore, the elimination of studies using valproic acid did not change the overall results (CC, P = 0.01; CT, P = 0.06; TT, P = 0.20) or subanalysis (CC, P = 0.23; CT, P = 0.13; TT, P = 0.73). In order to further ensure ethnic homogeneity in study subjects, we performed a meta-analysis excluding the studies from Malaysia. In this case, the overall (CC, P = 0.009; TT, P = 0.22) and subgroup (CC, P = 0.19; CT, P = 0.06; TT, P = 0.77) results did not change with the exception that the analysis on the CT genotype yielded a positive result (CT, P = 0.03). In summary, we found in a meta-analysis that epilepsy in Chinese patients with the MDR1 3435CC genotype has a higher risk of resistance to AEDs compared with epilepsy in those patients with other genotypes. Because of significant heterogeneity in the data evaluated and the lack of a correlation between genotype and drug resistance in the subgroup of patients from the southern region of China, additional studies with larger sample sizes and higher quality data to allow exclusion of the influence of drug type, region, ethnicity, age,
J.-W. Cheng et al. / Epilepsy & Behavior 36 (2014) 173–179
gender, testing methods, and diagnostic criteria are needed to confirm the conclusion. References [1] Sills GJ, Mohanraj R, Butler E, McCrindle S, Collier L, Wilson EA, et al. Lack of association between the C3435T polymorphism in the human multidrug resistance (MDR1) gene and response to antiepileptic drug treatment. Epilepsia 2005;46: 643–7. [2] Szoeke C, Sills GJ, Kwan P, Petrovski S, Newton M, Hitiris N, et al. Multidrug-resistant genotype (ABCB1) and seizure recurrence in newly treated epilepsy: data from international pharmacogenetic cohorts. Epilepsia 2009;50:1689–96. [3] Chang L, Wang XS. Investigation progress on the epidemiology of epilepsy in China. J Int Neurol Neurosurg 2012;39(2):161–4. [4] Siddiqui A, Kerb R, Weale ME, Brinkmann U, Smith A, Goldstein DB, et al. Association of multidrug resistance in epilepsy with a polymorphism in the drug-transporter gene ABCB1. N Engl J Med 2003;348:1442–8. [5] Nockel J, van den Engel NK, Winter H, Hatz RA, Zimmermann W, Kammerer R. Characterization of gastric adenocarcinoma cell lines established from CEA424/SV40 T antigen-transgenic mice with or without a human CEA transgene. BMC Cancer 2006;6:57. [6] Schmidt D, Loscher W. Drug resistance in epilepsy: putative neurobiologic and clinical mechanisms. Epilepsia 2005;46:858–77. [7] Calatozzolo C, Pollo B, Botturi A, Dinapoli L, Carosi M, Salmaggi A, et al. Multidrug resistance proteins expression in glioma patients with epilepsy. J Neurooncol 2012;110:129–35. [8] Komar AA. Silent SNPs: impact on gene function and phenotype. Pharmacogenomics 2007;8:1075–80. [9] Seo T, Ishitsu T, Ueda N, Nakada N, Yurube K, Ueda K, et al. ABCB1 polymorphisms influence the response to antiepileptic drugs in Japanese epilepsy patients. Pharmacogenomics 2006;7:551–61. [10] Kwan P, Wong V, Ng PW, Lui CH, Sin NC, Poon WS, et al. Gene-wide tagging study of association between ABCB1 polymorphisms and multidrug resistance in epilepsy in Han Chinese. Pharmacogenomics 2009;10:723–32. [11] Sayyah M, Kamgarpour F, Maleki M, Karimipoor M, Gharagozli K, Shamshiri AR. Association analysis of intractable epilepsy with C3435T and G2677T/A ABCB1 gene polymorphisms in Iranian patients. Epileptic Disord 2011;13:155–65. [12] Kwan P, Baum L, Wong V, Ng PW, Lui CH, Sin NC, et al. Association between ABCB1 C3435T polymorphism and drug-resistant epilepsy in Han Chinese. Epilepsy Behav 2007;11:112–7. [13] Tan NC, Heron SE, Scheffer IE, Pelekanos JT, McMahon JM, Vears DF, et al. Failure to confirm association of a polymorphism in ABCB1 with multidrug-resistant epilepsy. Neurology 2004;63:1090–2. [14] Shahwan A, Murphy K, Doherty C, Cavalleri GL, Muckian C, Dicker P, et al. The controversial association of ABCB1 polymorphisms in refractory epilepsy: an analysis of multiple SNPs in an Irish population. Epilepsy Res 2007;73:192–8. [15] Sterjev Z, Trencevska GK, Cvetkovska E, Petrov I, Kuzmanovski I, Ribarska JT, et al. The association of C3435T single-nucleotide polymorphism, Pgp-glycoprotein gene expression levels and carbamazepine maintenance dose in patients with epilepsy. Neuropsychiatr Dis Treat 2012;8:191–6. [16] Ameyaw MM, Regateiro F, Li T, Liu X, Tariq M, Mobarek A, et al. MDR1 pharmacogenetics: frequency of the C3435T mutation in exon 26 is significantly influenced by ethnicity. Pharmacogenetics 2001;11:217–21. [17] Kimchi-Sarfaty C, Marple AH, Shinar S, Kimchi AM, Scavo D, Roma MI, et al. Ethnicity-related polymorphisms and haplotypes in the human ABCB1 gene. Pharmacogenomics 2007;8:29–39. [18] Hung CC, Tai JJ, Lin CJ, Lee MJ, Liou HH. Complex haplotypic effects of the ABCB1 gene on epilepsy treatment response. Pharmacogenomics 2005;6:411–7. [19] Hung CC, Jen Tai J, Kao PJ, Lin MS, Liou HH. Association of polymorphisms in NR1I2 and ABCB1 genes with epilepsy treatment responses. Pharmacogenomics 2007;8:1151–8. [20] Haerian BS, Lim KS, Tan HJ, Mohamed EH, Tan CT, Raymond AA, et al. Association between ABCB1 polymorphism and response to sodium valproate treatment in Malaysian epilepsy patients. Epileptic Disord 2011;13:65–75. [21] Haerian BS, Lim KS, Mohamed EH, Tan HJ, Tan CT, Raymond AA, et al. Lack of association of ABCB1 and PXR polymorphisms with response to treatment in epilepsy. Seizure 2011;20:387–94. [22] Dong L, Luo R, Tong Y, Cai X, Mao M, Yu D. Lack of association between ABCB1 gene polymorphisms and pharmacoresistant epilepsy: an analysis in a western Chinese pediatric population. Brain Res 2011;1391:114–24. [23] Lu JC, Ren HM, Zhu GX. Association of the C3435T polymorphism in the multidrug resistance gene 1 and response to antiepileptic drug treatment in epilepsy patients. Chin J Neurol 2007;40(9):584–7. [24] Zheng XZ, Wu SJ, Xia M. Association of a polymorphism in MDR1 C3435T with response to antiepileptic drugs treatment inintractable epilepsy. J Apoplexy Nerv Dis 2009;26(1):44–6.
179
[25] Ding YY, Yang F, Liu J. Effects of the multidrug resistance 1 gene haplotype on resistance of valproic acid in Chinese Han Epilepsy patients. Chin J Clin Pharmacol Ther 2010;15(11):1267–71. [26] Dong T, Xu XR, Qi MS. Association analysis of ABCB1 C3435T polymorphism and drug -resistant epilepsy in Hui and Han-nation epilepsy patients. J Apoplexy Nerv Dis 2011;28(3):237–41. [27] Gao X, Zhou SZ, Guo Q. Polymorphism of MDR1 in childhood refractory epilepsy. J Clin Pediatr 2009;27(11):1014–8. [28] Wang SY, Gao F, Xia ZZ. Association of a polymorphism in MDR1 C3435T with intractable epilepsy of children. Zhejiang Med J 2008;30(12):1307–10. [29] Chen L, Liu CQ, Hu Y, Xiao ZT, Chen Y, Liao JX. Association of a polymorphism in MDR1 C3435T with response to antiepileptic drug treatment in ethic Han Chinese children with epilepsy. Chin J Contemp Pediatr 2007;9(1):11–4. [30] Guo GM, Guo YJ, Lin WH. Association study of C3435T genetic polymorphism of MDR1 gene with efficacy of carbamazepine. J Apoplexy Nerv Dis 2010;27(2): 128–30. [31] Jin RF, Yang CY, Xu XS. Frequency of the C3435T polymorphism in MDR1 genes and association of the polymorphisms with multidrug-resistant epilepsy. Chin J Clin 2010;4(11):2153–7. [32] Wang XY, Lin RF, Shi DH. Association of C3435T polymorphism in multidrug resistance gene 1 with refractory epilepsy. Prog Mod Biomed 2011;11(14):2717–9. [33] Di Q, Wang LL, Xu LG. Association between the C3435T polymorphism of human multidrug resistance 1 gene and refractory epilepsy. Chin J Neuromed 2011;10(2): 127–31. [34] Yang ML, Li T, Huang YH. The relationship between genetic polymorphism and expressed matter of MDR1 gene with efficacy of carbamazepine. Chin J Diffic Complicat Cases 2012;11(8):594–6. [35] Dong T, Xu XR, Zhang Q. Association between 1 C3435T and drug resistant epilepsy in Ningxia Hui epilepsy patients. J Ningxia Med Univ 2011;33(2):108–11. [36] Basic S, Hajnsek S, Bozina N, Filipcic I, Sporis D, Mislov D, et al. The influence of C3435T polymorphism of ABCB1 gene on penetration of phenobarbital across the blood-brain barrier in patients with generalized epilepsy. Seizure 2008;17: 524–30. [37] Lovric M, Bozina N, Hajnsek S, Kuzman MR, Sporis D, Lalic Z, et al. Association between lamotrigine concentrations and ABCB1 polymorphisms in patients with epilepsy. Ther Drug Monit 2012;34:518–25. [38] Loscher W, Klotz U, Zimprich F, Schmidt D. The clinical impact of pharmacogenetics on the treatment of epilepsy. Epilepsia 2009;50:1–23. [39] Soranzo N, Cavalleri GL, Weale ME, Wood NW, Depondt C, Marguerie R, et al. Identifying candidate causal variants responsible for altered activity of the ABCB1 multidrug resistance gene. Genome Res 2004;14:1333–44. [40] Lakhan R, Misra UK, Kalita J, Pradhan S, Gogtay NJ, Singh MK, et al. No association of ABCB1 polymorphisms with drug-refractory epilepsy in a north Indian population. Epilepsy Behav 2009;14:78–82. [41] Bournissen FG, Moretti ME, Juurlink DN, Koren G, Walker M, Finkelstein Y. Polymorphism of the MDR1/ABCB1 C3435T drug-transporter and resistance to anticonvulsant drugs: a meta-analysis. Epilepsia 2009;50:898–903. [42] Haerian BS, Roslan H, Raymond AA, Tan CT, Lim KS, Zulkifli SZ, et al. ABCB1 C3435T polymorphism and the risk of resistance to antiepileptic drugs in epilepsy: a systematic review and meta-analysis. Seizure 2010;19:339–46. [43] Wang P, Xia HH, Zhang JY, Dai LP, Xu XQ, Wang KJ. Association of interleukin-1 gene polymorphisms with gastric cancer: a meta-analysis. Int J Cancer 2007;120(3): 552–62. [44] Li P, Qin C. Methylenetetrahydrofolate reductase (MTHFR) gene polymorphisms and susceptibility to ischemic stroke: a meta-analysis. Gene 2014;535(2):359–64. [45] Marie SK, Shinjo SK, Oba-Shinjo SM, da Silva R, Barbosa KC, Yamamoto F, et al. Methylenetetrahydrofolate reductase gene polymorphism is not related to the risk of ischemic cerebrovascular disease in a Brazilian population. Clinics (Sao Paulo) 2007;62(3):295–300. [46] Baltes S, Fedrowitz M, Tortos CL, Potschka H, Löscher W. Valproic acid is not a substrate for p-glycoprotein or multidrug resistance proteins 1 and 2 in a number of in vitro and in vivo transport assays. J Pharmacol Exp Ther 2007;320:331–43. [47] Ufer M, Mosyagin I, Muhle H, Jacobsen T, Haenisch S, Häsler R. Non-response to antiepileptic pharmacotherapy is associated with the ABCC2–24CNT polymorphism in young and adult patients with epilepsy. Pharmacogenet Genomics 2009;19:353–62. [48] Sills GJ, Szoeke C, Kwan P, Petrovski S, Newton M, Hitiris N. Multidrug-resistant genotype (ABCB1) and seizure recurrence in newly treated epilepsy: data from international pharmacogenetic cohorts. Epilepsia 2009;50:1689–96. [49] Seo T, Ishitsu T, Ueda N, Nakada N, Yurube K, Ueda K. ABCB1 polymorphisms influence the response to antiepileptic drugs in Japanese epilepsy patients. Pharmacogenomics 2006;7:551–61.
Contents lists available at ScienceDirect
Epilepsy & Behavior journal homepage: www.elsevier.com/locate/yebeh
Association between MDR1 C3435T polymorphism and refractory epilepsy in the Chinese population: A systematic review and meta-analysis Ji-Wei Cheng a, Li-Jun Zhang a, Yu-Qing Hou a, Qing Zhao a, Xiao-Jing Zhang a, Xue-Fen Chen b, Yu Bai a,⁎ a b
Department of Neurology, Putuo Hospital, Shanghai University of Traditional Chinese Medicine, Shanghai 200062, China Department of Preventive Medicine and Health Statistics of Shanghai University of Traditional Chinese Medicine, Shanghai 201203, China
a r t i c l e
i n f o
Article history: Received 29 January 2014 Revised 8 April 2014 Accepted 6 May 2014 Available online xxxx Keywords: MDR1 C3435T Polymorphism Epilepsy Chinese Meta-analysis
a b s t r a c t The association between the C3435T polymorphism in the MDR1 gene and refractory epilepsy remains controversial. The association appears to be influenced by ethnicity and region. We have performed a systematic review and meta-analysis to assess the link between the MDR1 C3435T polymorphism and refractory epilepsy in the Chinese population. We searched the Cochrane Library, MIDLINE, EMBASE, CBM disc, CNKI, VIP, and WANFANG databases for literature published through August 2013 for case–control studies that evaluated the association between the MDR1 C3435T polymorphism and refractory epilepsy. Twenty-one case–control studies involving 4269 patients (1863 cases in the group with drug-resistant epilepsy and 2406 in the group with drug-responsive epilepsy) were included in the systematic review and metaanalysis. The analysis showed that there were significantly more cases with the MDR1 3435 CC genotype in the group with drug-resistant epilepsy than in the group with drug-responsive epilepsy [odds ratio (OR) = 1.50, 95% confidence interval (CI) = 1.09–2.06, P = 0.01]. In a subanalysis of patients from the southern regions of China, the correlation was not significant [odds ratio (OR) = 1.2, 95% confidence interval (CI) = 0.89–1.64, P = 0.24]. The relationship established in a subset of the Chinese population between the MDR1 C3435T polymorphism and refractory epilepsy will guide epilepsy treatment and development of new AEDs. © 2014 Elsevier Inc. All rights reserved.
1. Introduction Approximately 50 million people worldwide suffer from epilepsy, and roughly one-third cannot control the disease despite treatment with appropriate antiepileptic drugs (AEDs) [1,2]. China has about 9 million people with epilepsy [3]. A number of studies have demonstrated a relationship between refractory epilepsy and the multidrug resistance gene 1 (MDR1/ABCB1) [4–7]. MDR1 is highly polymorphic, and the C3435T polymorphism is one of the most common [8]. Although the C3435T polymorphism is associated with refractory epilepsy [4, 9–11], there are indications that the link is specific to only certain populations. The studies of Siddiqui et al. [4] and Sayyah et al. [11] found that patients with drug-resistant epilepsy were more likely to have a CC genotype at MDR1 3435 than a TT genotype. However, Seo et al. [9] and Kwan et al. [12] found that patients with drug-resistant epilepsy were more likely to have TT genotypes. Other studies [2,13–15] have found no relation between the MDR1 sequence and drug resistance. It has been suggested that these differences may be related to ethnicity
⁎ Corresponding author. Tel./fax: +86 13818606517. E-mail address: [email protected] (Y. Bai).
http://dx.doi.org/10.1016/j.yebeh.2014.05.007 1525-5050/© 2014 Elsevier Inc. All rights reserved.
and region [16,17]. We have retrieved relevant studies published in China and abroad and have carried out a systematic review and metaanalysis on the link between the C3435T polymorphism and refractory epilepsy in the Chinese population. 2. Methods 2.1. Search strategy and selection We searched the Cochrane Library, MIDLINE, EMBASE, CBM disc, CNKI, VIP, and WANFANG databases for case–control or cohort studies published through August of 2013 that evaluated the association between polymorphisms of MDR1 C3435T and refractory epilepsy in Chinese populations, region and language unlimited. The case–control studies included in this meta-analysis were required to meet the following criteria: (a) patients were treated with AEDs, and compliance was demonstrated; (b) data on genotype distributions were available for both case and control groups; (c) genotype distribution of the control subjects conformed to the Hardy–Weinberg equilibrium; and (d) there was a clear treatment outcome of either resistance or responsiveness to AEDs. The Boolean search strategy “C3435T” AND “[MDR1 OR ABCB1 OR P-glycoprotein (P-gp)]” AND “epilepsy” AND “Chinese” was
Various Various Various CBZ Various Various VPA VPA/CBZ Various Various Various
Various Various Various Various Various Various Various Various Various Various
76 178 61 93 12 152 20 33 58 35 T 59 27 45 33 101 84 86 115 111 57 22
Abbreviations: R, drug-responsive; NR, drug-resistant; MDR1, multidrug resistance 1. VPA, valproate; CBZ, carbamazepine; “various” indicates that the patient was treated with two or more drugs.
Sex (m) / / 32 / / / 33 43 54 32 Sex (m) / 25 / / 67 / 56 / 108 47 19 Age 38.5 ± 13.4 / 29.39 ± 14.71 / / / 27.9 ± 12.1 27 ± 13 37.33 ± 15.56 36.51 ± 13.50 Age / 5.3 ± 1.6 1.5–15 16–73 5–59 / 30 ± 19 / 6.6 ± 3.9 47.33 ± 20.39 21.2 ± 17.5 T 266 205 69 241 22 190 42 68 102 64 T 45 36 123 40 123 66 89 125 139 64 25 C 180 389 55 185 46 350 24 82 78 56 C 79 44 205 58 121 76 105 167 247 94 27 TT 74 22 16 67 1 21 15 12 29 19 TT 8 8 22 11 19 14 22 28 28 17 5 CT 118 161 37 107 20 148 12 44 44 26 CT 29 20 79 18 85 38 45 69 83 30 15
R group (genotype/allele/age/sex)
CC 31 114 9 39 13 101 6 19 17 15 CC 25 12 63 20 18 19 30 49 82 32 6 Sex (m) / / 51 / / / 19 31 41 30 Sex (m) / 23 / / 62 / 56 / 95 54 18 Age 39.1 ± 11 / 32.58 ± 15.41 / / / 33.8 ± 13.8 29 ± 12 36.74 ± 11.63 35.28 ± 13.20 Age / 5.7 ± 1.3 1.5–15 16–73 5–55 / 29 ± 18 / 6.2 ± 3.6 30.96 ± 15.18 31.3 ± 16.5 T
C 140 264 83 135 10 236 42 103 132 91 C 81 53 55 29 115 86 114 147 203 125 24 TT 15 37 13 19 2 29 9 10 5 6 TT 13 4 10 9 18 20 16 23 18 10 4 China (Shanghai) China (Zhejiang) China (Shenzhen/Hong Kong) China (Jilin) China (Shandong) China (Fujian) Malaya (Kuala Lumpur) Malaya (Kuala Lumpur) China (Sichuan) China (Nanjing) China (Shenzhen)
In seven of the 21 published articles, the authors concluded that patients with epilepsy with the MDR1 3435 CC genotype were more likely to have drug-resistant epilepsy compared with those with a different genotype. Three studies identified an association between the 3435 TT genotype and drug resistance; however, 11 studies found no such link. Our meta-analysis showed that there were more MDR1 3435 CC genotype patients in the group with drug-resistant epilepsy than in the group with drug-responsive epilepsy (Fig. 1 and Table 2). Our analysis indicated no significant association between the CT or TT genotypes and resistance to AEDs (Figs. 2 and 3 and Table 2). Studies included in
China (Taiwan) China (Hong Kong) China (Shanghai) China (Taiwan) China (Hong Kong) China (Hong Kong) China (Tianjin) China (Beijing) China (Ningxia) China (Ningxia)
3.2. Meta-analysis
CT 46 104 35 55 8 94 2 13 48 23 CT 33 19 25 15 65 44 54 69 75 32 14
Twenty-two published articles were retrieved that evaluated the relationship between the MDR1 polymorphisms and refractory epilepsy in Chinese patients. Of these articles, eight [2,10,12,18–22] were published in English, and 14 [23–35] were published in Chinese. One of these was a published dissertation (written in Chinese). All but the dissertation were included in the systematic review and meta-analysis. The results of each included study are summarized in Table 1. Six studies were conducted in northern China, 13 studies in southern China, and two in Malaysia (included in the South region for subanalysis). A total of 4269 patients were included in the meta-analysis; 1863 cases had drug-resistant epilepsy, and 2406 cases had drug-responsive epilepsy. There were no significant difference in ages (WMD = − 0.18, 95% CI = − 1.83 to 1.46, P = 0.83) or genders (OR = 0.98, 95% CI = 0.81 to 1.19, P = 0.86) between the two groups.
CC 47 80 24 40 1 71 20 45 42 34 CC 24 17 15 7 25 21 30 39 64 44 5
3.1. Study characteristics
NR group (genotype/allele/age/sex)
3. Results
Countries (districts)
The RevMan 4.2 software was used for the meta-analysis. The odds ratios (ORs) and 95% confidence intervals (CIs) of the binary classification variable data and the weighted mean differences (WMDs) and 95% CIs of the continuous variable data were calculated using a fixed effects model if there was no significant heterogeneity between each test (P N 0.05, I2 b 50%). If heterogeneity was present, a random effects model was employed. Twenty-one studies were analyzed. The study populations were divided for subanalyses into patients from two regions to explore the influence of region: the North (northern China) and the South (including southern China and Malaysia) as separated by the Yangtze River. We also performed a sensitivity analysis by eliminating each study in turn and a cumulative analysis by literature publication dates to test the stability of our conclusions and identify the sources of heterogeneity. Publication bias was evaluated by visual funnel plot inspection and Egger's test.
Positive studies (n = 10) Hung et al. [18] Kwan et al. [12] Lu et al. [23] Hung et al. [19] Szoeke et al. [2] Kwan et al. [10] Zheng et al. [24] Ding et al. [25] Dong et al. [22] Dong et al. [26] Negative studies (n = 11) Gao et al. [27] Wang et al. [28] Chen et al. [29] Guo et al. [30] Jin et al. [31] Wang et al. [32] Haerian et al. [20] Haerian et al. [21] Dong et al. [35] Di et al. [33] Yang et al. [34]
2.3. Statistical analysis
Table 1 Characteristics of the studies of the MDR1 C3435T polymorphism and the distribution of the MDR1 C3435T genotypes and allele frequencies among patients whose epilepsy is sensitive (NR) or resistant (R) to AEDs.
We downloaded and printed the full text of all relevant publications. Two trained valuators read the literature and extracted the relevant information into registration forms according to a standardized method. Each publication was read and data were extracted by both valuators. The information included the name, gender, and age of first author of the study, year of publication, sample size, the country (district) of the study, and the research conclusions. Disagreements were documented and resolved by discussion by a third author.
Drugs
2.2. Data extraction
CC TT CC CC TT TT CC CC CC CC
Frequent genotype in NR
used. We also retrieved the related references listed in the studies included in our meta-analysis. All studies included were published literature; dissertations and conference papers were not included.
– – – – – – – – – – –
J.-W. Cheng et al. / Epilepsy & Behavior 36 (2014) 173–179
Studies
174
J.-W. Cheng et al. / Epilepsy & Behavior 36 (2014) 173–179
175
Fig. 1. Forest plot of MDR1 CC genotype frequencies in meta-analysis of patients with drug-resistant and drug-responsive epilepsies.
the meta-analysis were heterogeneous, so we adopted the random effects model. Visual funnel plot inspection and Egger's test showed no obvious publication biases (CC, P = 0.49; CT, P = 0.64; TT, P = 0.80). We conducted a region-based subanalysis, sensitivity analysis, and cumulative meta-analysis by literature publication date to test the stability of our conclusions and to identify the sources of heterogeneity. The subanalysis showed that in patients from the southern region of China, which included Malaysia, there was no significant association between the MDR1 3435 CC genotype and drug resistance (Fig. 1 and Table 2). The subanalysis of the patients from the North was consistent with the overall results that the 3435 CC genotype was correlated with drug resistance (Table 2).
We performed a sensitivity analysis by eliminating each study in turn. The heterogeneity of the North subanalysis on 3435 CC disappeared in the absence of the study of Guo et al. [30]. The heterogeneity of the North subanalysis and of the total analysis on 3435 CT decreased when data from the study of Ding et al. [25] were removed. After eliminating the study of Dong et al. [26], the heterogeneity of the North subanalysis on 3435 TT disappeared. After eliminating the studies of Hung et al. [18,19], the heterogeneity of the South subanalysis on 3435 CC disappeared. We also conducted a cumulative meta-analysis on the association between the MDR1 3435 CC genotype and drug resistance by literature publication date; the results were in agreement with those obtained when all data were analyzed (Fig. 4).
Table 2 Distribution of MDR1 C3435T genotype frequencies among patients with drug-resistant and drug-responsive epilepsies based on district within China. Genotypes
Studies
CC
T N S T N S T N S
TT
CT
n/N
Fixed-effects model
NR
R
695/1863 173/396 522/1467 290/1863 57/396 233/1467 873/1863 166/396 707/1467
720/2406 95/429 625/1977 458/2406 105/429 353/1977 1228/2406 229/429 999/1977
P
0.22
OR (95% CI)
0.92 (0.80–1.05)
Random-effects model P
OR (95% CI)
0.01 0.01 0.24 0.16 0.10 0.61 0.10 0.10
1.50 (1.09–2.06) 2.75 (1.36–5.55) 1.20 (0.89–1.64) 0.78 (0.54–1.13) 0.52 (0.24–1.13) 0.91 (0.62–1.32) 0.85 (0.71–1.03) 0.57 (0.29–1.11)
I2 (%)
P
79.90 78.30 73.40 73.00 75.40 71.70 49.50 78.70 0.00
0.00 0.00 0.00 0.00 0.00 0.00 0.01 0.00 0.64
Abbreviations: R, drug-responsive; NR, drug-resistant; T, includes all studies; N, includes only studies on northern China; S, includes studies on southern China and Malaysia.
176
J.-W. Cheng et al. / Epilepsy & Behavior 36 (2014) 173–179
Fig. 2. Forest plot of MDR1 CT genotype frequencies in meta-analysis of patients with drug-resistant and drug-responsive epilepsies.
4. Discussion Drug resistance is mainly mediated by the protein products of the MDR1 and MDR2 genes. P-glycoprotein, which is the expression product of MDR1, is expressed in the intestine, placenta, blood–brain barrier, kidney, and liver. The substrates of P-gp include liposoluble drugs. Most AEDs, including phenytoin, phenobarbital, and lamotrigine, are liposoluble. In patients whose epilepsy is resistant to AEDs, it is assumed that P-gp actively transports these AEDs from cells [36,37]. In humans, MDR1 is located on chromosome 7 q21.1, and more than 50 single nucleotide polymorphisms (SNPs) have been identified. There is an association between C3435T of exon 26 of the MDR1 gene and the expression and function of P-gp [38]. Siddiqui et al. [4] were first to report that patients with drugresistant epilepsy were more likely to have the CC genotype at MDR1 3435 than the TT genotype. Results of subsequent studies have been inconsistent. Data from some studies [11,39] have supported this association, whereas others [18,19] found that patients with the TT genotype were more likely to have drug-resistant epilepsy, and, still, other studies [2,13–15,21,22,40] failed to find any association between this position of MDR1 and drug resistance in patients with epilepsy. There are a number of potential causes for these contradictory results. First, a uniform definition of refractory epilepsy was not used in previous analyses, making division of patients into those with drug-resistant and those with drug-responsive epilepsy difficult [1,4]. As listed in Table 3, different definitions of drug resistance were used in the studies analyzed in our meta-analysis. In the studies of Hung et al. [18,19], drug
resistance was defined as 10 seizures in a year with two or more AEDs at maximally tolerated doses. Kwan et al. [12] defined it as one seizure in a month with two or more AEDs at therapeutic dosages, and Haerian et al. [21] defined it as one seizure in a year with valproate or carbamazepine monotherapy at maximally tolerated dosages. In contrast, most of the studies of Chinese mainland subjects [23–26,28,30–32,34,35] defined drug resistance as less than 50% seizure reduction with AEDs at maximally tolerated doses or more than four seizures in a month with appropriate first-line AEDs at therapeutic dosages for 2 years. These differences made it difficult to group patients according to the same criterion. Second, the types and doses of AEDs vary from study to study. Certain antiepileptic drugs, such as phenobarbital, are P-gp substrates, whereas others, such as sodium valproate, are not. Other drugs that might be taken by patients with epilepsy, including calcium channel blockers and antibiotics, are also P-gp substrates and might have confounded results. Third, the frequency of the C3435T polymorphism varies with ethnicity in healthy subjects. Those of African descent are most likely to have the 3435 CC genotype with Europeans least likely [16,17,41]. Fourth, age and gender may impact results. The study of Sanchez et al. on adults with epilepsy found that 3435 CC genotype patients were more likely to have drug-resistant epilepsy compared with those with a different genotype; however, this was not the case for children. Sayyah et al. [11] found that female, but not male, patients in Iran with the 3435 CC genotype were more likely to have drug-resistant epilepsy compared with those with other genotypes. Fifth, differences in genotype analysis methods may also affect the results: Lakhan et al. [40] detected a 231-bp PCR product using DpnII restriction endonuclease,
J.-W. Cheng et al. / Epilepsy & Behavior 36 (2014) 173–179
177
Fig. 3. Forest plot of MDR1 TT genotype frequencies in meta-analysis of patients with drug-resistant and drug-responsive epilepsies.
Fig. 4. Cumulative meta-analysis of the relationship between MDR1 3435 CC and drug resistance by study publication date.
whereas Lu et al. [23] detected a 248-bp PCR product using DpnI. Finally, the etiopathogenesis of refractory epilepsy is complex. It appears to be associated with excessive expression of P-gp in the brain and may be affected by CY450, interleukin, and other enzymes that alter the metabolism of AEDs. Meta-analysis can increase the reliability of conclusions by increasing sample size. Data from 21 studies in which drug resistance or sensitivity and genotypes of patients with epilepsy from China were determined were included in our meta-analysis. These case–control studies involved 4269 patients. The meta-analysis showed that a significantly higher fraction of patients in the group with drug-resistant epilepsy than in the group with drug-sensitive epilepsy had the MDR1 3435 CC genotype. However, two previous meta-analyses did not support the association [41,42]. In the study of Bournissen et al. [41], 11 studies from six countries around the world involving 3371 patients were included in the analysis. We found four positive and seven negative studies in that analysis. The most frequent genotype in the group with drug-resistant epilepsy from only one of these studies was 3435 CC. The meta-analysis found no association between the ABC1 genotype and response to AEDs. In the meta-analysis of Haerian et al. [42], 22 studies from 11 countries around the world involving 6755 patients were included in the analysis. There were nine positive and 13 negative studies in the analysis. The most frequent genotype in the group with drug-resistant epilepsy from five of the studies was 3435 CC and
178
J.-W. Cheng et al. / Epilepsy & Behavior 36 (2014) 173–179
Table 3 Definitions of drug responsiveness and drug resistance used in analyzed studies. Studies
Countries (districts)
Definition of drug responsiveness and drug resistance Drug responsiveness
Drug resistance N10 seizures in a year with 2 or more AEDs at maximally tolerated doses N1 seizure in a month with 2 or more AEDs at therapeutic dosages
Positive studies Hung et al. [18] Kwan et al. [12]
China (Taiwan) China (Hong Kong)
Lu et al. [23] Hung et al. [19] Szoeke et al. [2] Kwan et al. [10] Zheng et al. [24]
China (Shanghai) China (Taiwan) China (Hong Kong) China (Hong Kong) China (Tianjin)
N2 years seizure freedom N1 year seizure freedom with a stable dose of an AED ≥50% seizure reduction As Hung et al. [18] N1 year seizure freedom As Kwan et al. [12] N1 year seizure freedom
Ding et al. [25] Dong et al. [22] Dong et al. [26]
China (Beijing) China (Ningxia) China (Ningxia)
≥50% seizure reduction ≥50% seizure reduction ≥50% seizure reduction
Negative studies Gao et al. [27]
China (Shanghai)
≥1 year seizure freedom
Wang et al. [28]
China (Zhejiang)
N1 year seizure freedom
Chen et al. [29] Guo et al. [30]
China (Shenzhen/Hong Kong) China (Jilin)
Jin et al. [31]
China (Shandong)
Wang et al. [32]
China (Fujian)
N1 year seizure freedom ≥50% seizure reduction with CBZ monotherapy N1 year seizure freedom or ≥50% seizure reduction N1 year seizure freedom
Haerian et al. [20]
Malaya (Kuala Lumpur)
Haerian et al. [21]
Malaya (Kuala Lumpur)
Dong et al. [35] Di et al. [33] Yang et al. [34]
China (Sichuan) China (Nanjing) China (Shenzhen)
N1 year seizure freedom with VPA treatment N1 year seizure freedom with CBZ or VPA monotherapy N1 year seizure freedom N1 year seizure freedom N1 year seizure freedom
3435TT from four of the studies. The meta-analysis did not find an association between the ABC1 C3435T polymorphism and the risk of resistance to AEDs. The main difference between the two previous meta-analyses and the present study is the case ethnicities. Our study concentrated on a Chinese population, whereas cases analyzed in the previous studies were from around the world. There is precedence for our finding that ethnicity is a factor in genetic linkage of diseases. Wang et al. [43] performed a meta-analysis with ethnically heterogeneous groups from all over the world including thirty-nine studies that included 6863 cases with gastric cancer and 8434 controls. The analysis suggested that the IL-1RN genetic polymorphisms were associated with an increased risk of developing gastric cancer. A subanalysis of the Chinese population did not find such an association. The meta-analysis performed by Li et al. [44] of nineteen case–control studies involving 2223 patients and 2936 controls suggested that the MTHFR C667T genetic polymorphism was significantly associated with increased risk of ischemic stroke. In contrast, the study of Marie et al. [45] found that the MTHFR gene polymorphism was not related to the risk of ischemic cerebrovascular disease in the Brazilian population. In order to test the stability of our conclusion, we divided our studies into subgroups of patients from the north of China to the south of China (including Malaysia) and conducted a cumulative meta-analysis by literature publication date. The results, with the exception of subanalysis of the data from the South, were much the same as those obtained when all data were analyzed simultaneously. Heterogeneity inspection must be conducted in meta-analysis as only the statistics of homogeneous data are capable of a merger. The heterogeneity of data used here was examined by χ2 and I2 tests. These tests suggested
b50% of seizure reduction with AEDs at maximally tolerated doses As Hung et al. [18] Seizure in a year As Kwan et al. [12] N4 seizures in a month with appropriate first-line AEDs at therapeutic dosages for 2 years b50% seizure reduction with AEDs at maximally tolerated doses b50% seizure reduction with AEDs at maximally tolerated doses b50% seizure reduction with AEDs at maximally tolerated doses N1 seizure in a month with 2 or more first-line AEDs at therapeutic dosages for 6 months N4 seizures in a month with appropriate first-line AEDs at therapeutic dosages for 2 years N4 seizures in a year b50% seizure reduction with CBZ monotherapy b50% seizure reduction with 2 or more AEDs at therapeutic dosages N4 seizures in a month with appropriate first-line AEDs at therapeutic dosages for 2 years N1 seizure in a year with VPA at maximally tolerated doses N1 seizure in a year with CBZ or VPA monotherapy at maximally tolerated dosages N4 seizures in a year ≥1 seizure in a month with 2 or more appropriate AEDs for 2 years N4 seizures in a month with appropriate first-line AEDs at therapeutic dosages for 2 years
heterogeneity, and, therefore, we adopted the random effects model. We also eliminated data from each study in turn to find the source of the heterogeneity. The heterogeneity decreased upon elimination of certain studies, but the removal of data from individual studies had no impact on the overall conclusion. Heterogeneity may be associated with differences in sample size. Visual funnel plot inspection and Egger's test did not show obvious publication bias. It is commonly believed that valproic acid is not a substrate of p-glycoprotein but a substrate of multidrug resistance-associated protein [46]. However, VPA was the most commonly used firstline AED in reports that demonstrated both negative [20,21,47,48] and positive [49] associations between polymorphisms and drug efficacy. Furthermore, the elimination of studies using valproic acid did not change the overall results (CC, P = 0.01; CT, P = 0.06; TT, P = 0.20) or subanalysis (CC, P = 0.23; CT, P = 0.13; TT, P = 0.73). In order to further ensure ethnic homogeneity in study subjects, we performed a meta-analysis excluding the studies from Malaysia. In this case, the overall (CC, P = 0.009; TT, P = 0.22) and subgroup (CC, P = 0.19; CT, P = 0.06; TT, P = 0.77) results did not change with the exception that the analysis on the CT genotype yielded a positive result (CT, P = 0.03). In summary, we found in a meta-analysis that epilepsy in Chinese patients with the MDR1 3435CC genotype has a higher risk of resistance to AEDs compared with epilepsy in those patients with other genotypes. Because of significant heterogeneity in the data evaluated and the lack of a correlation between genotype and drug resistance in the subgroup of patients from the southern region of China, additional studies with larger sample sizes and higher quality data to allow exclusion of the influence of drug type, region, ethnicity, age,
J.-W. Cheng et al. / Epilepsy & Behavior 36 (2014) 173–179
gender, testing methods, and diagnostic criteria are needed to confirm the conclusion. References [1] Sills GJ, Mohanraj R, Butler E, McCrindle S, Collier L, Wilson EA, et al. Lack of association between the C3435T polymorphism in the human multidrug resistance (MDR1) gene and response to antiepileptic drug treatment. Epilepsia 2005;46: 643–7. [2] Szoeke C, Sills GJ, Kwan P, Petrovski S, Newton M, Hitiris N, et al. Multidrug-resistant genotype (ABCB1) and seizure recurrence in newly treated epilepsy: data from international pharmacogenetic cohorts. Epilepsia 2009;50:1689–96. [3] Chang L, Wang XS. Investigation progress on the epidemiology of epilepsy in China. J Int Neurol Neurosurg 2012;39(2):161–4. [4] Siddiqui A, Kerb R, Weale ME, Brinkmann U, Smith A, Goldstein DB, et al. Association of multidrug resistance in epilepsy with a polymorphism in the drug-transporter gene ABCB1. N Engl J Med 2003;348:1442–8. [5] Nockel J, van den Engel NK, Winter H, Hatz RA, Zimmermann W, Kammerer R. Characterization of gastric adenocarcinoma cell lines established from CEA424/SV40 T antigen-transgenic mice with or without a human CEA transgene. BMC Cancer 2006;6:57. [6] Schmidt D, Loscher W. Drug resistance in epilepsy: putative neurobiologic and clinical mechanisms. Epilepsia 2005;46:858–77. [7] Calatozzolo C, Pollo B, Botturi A, Dinapoli L, Carosi M, Salmaggi A, et al. Multidrug resistance proteins expression in glioma patients with epilepsy. J Neurooncol 2012;110:129–35. [8] Komar AA. Silent SNPs: impact on gene function and phenotype. Pharmacogenomics 2007;8:1075–80. [9] Seo T, Ishitsu T, Ueda N, Nakada N, Yurube K, Ueda K, et al. ABCB1 polymorphisms influence the response to antiepileptic drugs in Japanese epilepsy patients. Pharmacogenomics 2006;7:551–61. [10] Kwan P, Wong V, Ng PW, Lui CH, Sin NC, Poon WS, et al. Gene-wide tagging study of association between ABCB1 polymorphisms and multidrug resistance in epilepsy in Han Chinese. Pharmacogenomics 2009;10:723–32. [11] Sayyah M, Kamgarpour F, Maleki M, Karimipoor M, Gharagozli K, Shamshiri AR. Association analysis of intractable epilepsy with C3435T and G2677T/A ABCB1 gene polymorphisms in Iranian patients. Epileptic Disord 2011;13:155–65. [12] Kwan P, Baum L, Wong V, Ng PW, Lui CH, Sin NC, et al. Association between ABCB1 C3435T polymorphism and drug-resistant epilepsy in Han Chinese. Epilepsy Behav 2007;11:112–7. [13] Tan NC, Heron SE, Scheffer IE, Pelekanos JT, McMahon JM, Vears DF, et al. Failure to confirm association of a polymorphism in ABCB1 with multidrug-resistant epilepsy. Neurology 2004;63:1090–2. [14] Shahwan A, Murphy K, Doherty C, Cavalleri GL, Muckian C, Dicker P, et al. The controversial association of ABCB1 polymorphisms in refractory epilepsy: an analysis of multiple SNPs in an Irish population. Epilepsy Res 2007;73:192–8. [15] Sterjev Z, Trencevska GK, Cvetkovska E, Petrov I, Kuzmanovski I, Ribarska JT, et al. The association of C3435T single-nucleotide polymorphism, Pgp-glycoprotein gene expression levels and carbamazepine maintenance dose in patients with epilepsy. Neuropsychiatr Dis Treat 2012;8:191–6. [16] Ameyaw MM, Regateiro F, Li T, Liu X, Tariq M, Mobarek A, et al. MDR1 pharmacogenetics: frequency of the C3435T mutation in exon 26 is significantly influenced by ethnicity. Pharmacogenetics 2001;11:217–21. [17] Kimchi-Sarfaty C, Marple AH, Shinar S, Kimchi AM, Scavo D, Roma MI, et al. Ethnicity-related polymorphisms and haplotypes in the human ABCB1 gene. Pharmacogenomics 2007;8:29–39. [18] Hung CC, Tai JJ, Lin CJ, Lee MJ, Liou HH. Complex haplotypic effects of the ABCB1 gene on epilepsy treatment response. Pharmacogenomics 2005;6:411–7. [19] Hung CC, Jen Tai J, Kao PJ, Lin MS, Liou HH. Association of polymorphisms in NR1I2 and ABCB1 genes with epilepsy treatment responses. Pharmacogenomics 2007;8:1151–8. [20] Haerian BS, Lim KS, Tan HJ, Mohamed EH, Tan CT, Raymond AA, et al. Association between ABCB1 polymorphism and response to sodium valproate treatment in Malaysian epilepsy patients. Epileptic Disord 2011;13:65–75. [21] Haerian BS, Lim KS, Mohamed EH, Tan HJ, Tan CT, Raymond AA, et al. Lack of association of ABCB1 and PXR polymorphisms with response to treatment in epilepsy. Seizure 2011;20:387–94. [22] Dong L, Luo R, Tong Y, Cai X, Mao M, Yu D. Lack of association between ABCB1 gene polymorphisms and pharmacoresistant epilepsy: an analysis in a western Chinese pediatric population. Brain Res 2011;1391:114–24. [23] Lu JC, Ren HM, Zhu GX. Association of the C3435T polymorphism in the multidrug resistance gene 1 and response to antiepileptic drug treatment in epilepsy patients. Chin J Neurol 2007;40(9):584–7. [24] Zheng XZ, Wu SJ, Xia M. Association of a polymorphism in MDR1 C3435T with response to antiepileptic drugs treatment inintractable epilepsy. J Apoplexy Nerv Dis 2009;26(1):44–6.
179
[25] Ding YY, Yang F, Liu J. Effects of the multidrug resistance 1 gene haplotype on resistance of valproic acid in Chinese Han Epilepsy patients. Chin J Clin Pharmacol Ther 2010;15(11):1267–71. [26] Dong T, Xu XR, Qi MS. Association analysis of ABCB1 C3435T polymorphism and drug -resistant epilepsy in Hui and Han-nation epilepsy patients. J Apoplexy Nerv Dis 2011;28(3):237–41. [27] Gao X, Zhou SZ, Guo Q. Polymorphism of MDR1 in childhood refractory epilepsy. J Clin Pediatr 2009;27(11):1014–8. [28] Wang SY, Gao F, Xia ZZ. Association of a polymorphism in MDR1 C3435T with intractable epilepsy of children. Zhejiang Med J 2008;30(12):1307–10. [29] Chen L, Liu CQ, Hu Y, Xiao ZT, Chen Y, Liao JX. Association of a polymorphism in MDR1 C3435T with response to antiepileptic drug treatment in ethic Han Chinese children with epilepsy. Chin J Contemp Pediatr 2007;9(1):11–4. [30] Guo GM, Guo YJ, Lin WH. Association study of C3435T genetic polymorphism of MDR1 gene with efficacy of carbamazepine. J Apoplexy Nerv Dis 2010;27(2): 128–30. [31] Jin RF, Yang CY, Xu XS. Frequency of the C3435T polymorphism in MDR1 genes and association of the polymorphisms with multidrug-resistant epilepsy. Chin J Clin 2010;4(11):2153–7. [32] Wang XY, Lin RF, Shi DH. Association of C3435T polymorphism in multidrug resistance gene 1 with refractory epilepsy. Prog Mod Biomed 2011;11(14):2717–9. [33] Di Q, Wang LL, Xu LG. Association between the C3435T polymorphism of human multidrug resistance 1 gene and refractory epilepsy. Chin J Neuromed 2011;10(2): 127–31. [34] Yang ML, Li T, Huang YH. The relationship between genetic polymorphism and expressed matter of MDR1 gene with efficacy of carbamazepine. Chin J Diffic Complicat Cases 2012;11(8):594–6. [35] Dong T, Xu XR, Zhang Q. Association between 1 C3435T and drug resistant epilepsy in Ningxia Hui epilepsy patients. J Ningxia Med Univ 2011;33(2):108–11. [36] Basic S, Hajnsek S, Bozina N, Filipcic I, Sporis D, Mislov D, et al. The influence of C3435T polymorphism of ABCB1 gene on penetration of phenobarbital across the blood-brain barrier in patients with generalized epilepsy. Seizure 2008;17: 524–30. [37] Lovric M, Bozina N, Hajnsek S, Kuzman MR, Sporis D, Lalic Z, et al. Association between lamotrigine concentrations and ABCB1 polymorphisms in patients with epilepsy. Ther Drug Monit 2012;34:518–25. [38] Loscher W, Klotz U, Zimprich F, Schmidt D. The clinical impact of pharmacogenetics on the treatment of epilepsy. Epilepsia 2009;50:1–23. [39] Soranzo N, Cavalleri GL, Weale ME, Wood NW, Depondt C, Marguerie R, et al. Identifying candidate causal variants responsible for altered activity of the ABCB1 multidrug resistance gene. Genome Res 2004;14:1333–44. [40] Lakhan R, Misra UK, Kalita J, Pradhan S, Gogtay NJ, Singh MK, et al. No association of ABCB1 polymorphisms with drug-refractory epilepsy in a north Indian population. Epilepsy Behav 2009;14:78–82. [41] Bournissen FG, Moretti ME, Juurlink DN, Koren G, Walker M, Finkelstein Y. Polymorphism of the MDR1/ABCB1 C3435T drug-transporter and resistance to anticonvulsant drugs: a meta-analysis. Epilepsia 2009;50:898–903. [42] Haerian BS, Roslan H, Raymond AA, Tan CT, Lim KS, Zulkifli SZ, et al. ABCB1 C3435T polymorphism and the risk of resistance to antiepileptic drugs in epilepsy: a systematic review and meta-analysis. Seizure 2010;19:339–46. [43] Wang P, Xia HH, Zhang JY, Dai LP, Xu XQ, Wang KJ. Association of interleukin-1 gene polymorphisms with gastric cancer: a meta-analysis. Int J Cancer 2007;120(3): 552–62. [44] Li P, Qin C. Methylenetetrahydrofolate reductase (MTHFR) gene polymorphisms and susceptibility to ischemic stroke: a meta-analysis. Gene 2014;535(2):359–64. [45] Marie SK, Shinjo SK, Oba-Shinjo SM, da Silva R, Barbosa KC, Yamamoto F, et al. Methylenetetrahydrofolate reductase gene polymorphism is not related to the risk of ischemic cerebrovascular disease in a Brazilian population. Clinics (Sao Paulo) 2007;62(3):295–300. [46] Baltes S, Fedrowitz M, Tortos CL, Potschka H, Löscher W. Valproic acid is not a substrate for p-glycoprotein or multidrug resistance proteins 1 and 2 in a number of in vitro and in vivo transport assays. J Pharmacol Exp Ther 2007;320:331–43. [47] Ufer M, Mosyagin I, Muhle H, Jacobsen T, Haenisch S, Häsler R. Non-response to antiepileptic pharmacotherapy is associated with the ABCC2–24CNT polymorphism in young and adult patients with epilepsy. Pharmacogenet Genomics 2009;19:353–62. [48] Sills GJ, Szoeke C, Kwan P, Petrovski S, Newton M, Hitiris N. Multidrug-resistant genotype (ABCB1) and seizure recurrence in newly treated epilepsy: data from international pharmacogenetic cohorts. Epilepsia 2009;50:1689–96. [49] Seo T, Ishitsu T, Ueda N, Nakada N, Yurube K, Ueda K. ABCB1 polymorphisms influence the response to antiepileptic drugs in Japanese epilepsy patients. Pharmacogenomics 2006;7:551–61.